[ARFC] wstETH and weETH E-Modes and LT/LTV Adjustments on Ethereum, Arbitrum, Base - 03.12.25

[ChaosLabs]

Summary

Chaos Labs proposes establishing the addition of the following E-Modes:

• weETH/WETH Liquid E-Mode on Arbitrum, Base, and Core Instances.
• wstETH/WETH Liquid E-Mode on Arbitrum, Base, and Core Instances.

We also propose adjusting weETH’s LTV on Base and Arbitrum to enhance user capital efficiency.

weETH

After carefully observing weETH’s liquidity and volatility across multiple chains, we recommend a series of changes to improve the asset’s capital efficiency.

Below, we present weETH’s liquidity across Base and Arbitrum over the past six months.

weETH’s liquidity on Arbitrum has shown a strong reliance over the past six months, steadily increasing despite the recent price changes. Currently, selling 500 weETH incurs less than 2% price slippage.

weETH’s liquidity on Base has remained similarly stable following an increase in the 4th quarter of 2024. Currently, selling 650 weETH incurs less than 2% price slippage.

Peg Stability

Below, we analyze weETH’s peg stability during the August 5, 2024, market stress event, in which WETH fell 21%. Evaluating weETH’s exchange rate stability is critical for assessing LTV and E-mode parameters, as price deviations between weETH and WETH can lead to delays in liquidation and arbitrage-driven borrowing, impacting risk exposure.

On August 5, the weETH/WETH market rate followed a similar trajectory, with a maximum deviation reaching 140 bps. While the recovery speed on both chains was comparable, Ethereum exhibited a slightly faster reversion to the internal exchange rate. On Arbitrum, the market rate took longer to revert compared to Base and Ethereum. However, all three instances demonstrated a successful recovery and quick mean reversion.

LT & LTV

Based on the above observations, weETH’s liquidity on Base has remained stable after recent improvements, and its ability to maintain a stable peg under stressful market conditions supports an LTV&LT increase. Therefore, we recommend increasing LTV to 75% and LT to 77%. Additionally, we propose establishing a separate weETH/WETH Liquid E-Mode with higher parameters to enhance users’ capital efficiency rather than adjusting the parameters within the existing E-Mode.

On Arbitrum, weETH demonstrates similar stability in liquidity compared to Base but exhibits slightly weaker peg stability than on Base and Ethereum. However, its properties are still sufficient to recommend increasing its collateral parameters to 75% and 77%, aligning them with those on Base.

However, given the reduced liquidity present on the Base chain we recommend adjusting the Liquidation Bonus within the weETH/WETH E-Mode to 1.25% to mitigate the price impact caused by large highly correlated positions.

Additionally, in order to provide the DAO with more granularity during future adjustments, we recommend the creation of a weETH/WETH Liquid E-Mode on Arbitrum and Core Instances such that its parameters are aligned with the currently active ETH-Correlated E-Mode on the respective chains.

wstETH

Following the introduction to Liquid E-Modes with Aave v3.2, which enables targeted combinations of collateral and debt assets, we propose aligning the wstETH’s more aggressive parameters currently present on the Prime instance across all Aave deployments. Thanks to wstETH’s battle-tested stability, this change will improve capital efficiency while maintaining reduced risk.

To align the LT and LTV of wstETH to those used within Prime’s E-Mode, we cover the asset’s volatility and peg over multiple instances.

On Base, wstETH’s liquidity against WETH has declined in recent months. However, thanks to the fast bridging speed and quick mean reversion properties, it is still sufficient to support the proposed parameters.

Its liquidity has also fallen slightly on Arbitrum, though it has been more stable in the past two months, with a 1,000 wstETH for WETH swap able to be completed for less than 1% price slippage.

Peg Stability

Price deviations can cause problems like arbitrage exploitation, where, given the underlying liquidation threshold, users take advantage of price differences between the protocol’s fundamental valuation of an asset and its market price, potentially destabilizing liquidity. This can lead to WETH debt accrual and thus, in tandem with WETH collateral liquidations and withdrawals due to the underwriting of temporary mispriced debt, lead to interest rate and utilization spikes, with LST/LRT-collateralized e-mode LTVs gradually (net) increasing while WETH collateralized stablecoin debt positions risk unperformed liquidations due to such utilization rates.

These deviations typically arise during periods of market stress. In such scenarios, deviations are driven by complex factors, including investor de-risking from volatile assets, liquidation cascades triggered by non-correlated positions such as stablecoin borrows against WETH, and, in rare cases, slashing events that erode confidence in an asset. The speed of reversion is generally contingent on the market priced implied duration risk associated with such a WETH-correlated asset, which can be dynamic dictated by exit queue considerations, as well as the game-theoretic value given by the fundamental robustness and on-chain liquidity of the asset itself.

For this reason, to gauge the resilience of wstETH’s peg to ETH across different chains, given by the available liquidity and leverage on the chain, we analyze price behavior during the market drop that occurred on August 5, 2024, during which WETH suffered a 21% daily drop, with the rest of the market following suit and LST/LRT assets showing significant deviations. Despite that, wstETH demonstrated strong reversion characteristics across Base, Arbitrum, and Scroll. This observation indicates a high degree of stability that supports the proposed E-Mode parameters.

• Ethereum: On Ethereum, wstETH dropped to a minimum of 1.1531 WETH, representing a 1.85% depeg. The asset quickly returned within 0.5% from its exchange rate, showing a strong mean reversion tendency.

• Base: On Base, wstETH dropped to a minimum of 1.1490 WETH, representing a 2.1% depeg. Despite this dip, the asset quickly returned to its peg, demonstrating resilience and a tendency toward mean reversion under stress.

• Arbitrum: On Arbitrum, wstETH experienced a smaller depeg of 1.66%, reaching a low of 1.1550 WETH. Although this recovery was slightly slower than on Base, wstETH still demonstrated consistent mean reversion and stability.

LT and LTV

The combination of robust liquidity, strong peg stability, and the inherent resilience of wstETH ensures that aligning LTV and LT parameters for the wstETH/WETH E-Mode on Arbitrum, Base and Core to the one on Lido instance is both feasible and prudent. This is because the ETH-correlated E-Mode configuration is constrained by the least liquid and least stable assets within the pool, limiting capital efficiency across the board. However, with these updated configurations, we can adopt parameters that better reflect the liquidity and stability of wstETH, allowing for higher capital efficiency in isolated E-Modes.

Hence, we recommend setting the LTV at 93.5% and the LT at 95.5%. By enabling a high level of leverage, these parameters are also expected to increase demand for WETH across the chains.

Collateral and Borrowable

To prevent same-asset looping, which could undermine future use of liquidity incentives and lead to unintended utilization, we recommend configuring wstETH as collateral and WETH as borrowable in the proposed E-Mode. This structure aligns with the approach detailed in this post, which advocates for selective collateral and borrowing configurations to maintain market efficiency.

Specification

Based on the observations and considerations outlined above, we recommend implementing the following liquid E-Mode configuration for the weETH/WETH pair on Arbitrum, Base, and Core Instances and proposing an LTV adjustment for weETH on both Arbitrum, Base, and Core Instances.

weETH/WETH E-Mode (Base)

Parameter	Value	Value
Asset	weETH	WETH
Collateral	Yes	No
Borrowable	No	Yes
Max LTV	93%	-
Liquidation Threshold	95%	-
Liquidation Penalty	1.25%	-

weETH/WETH E-Mode (Ethereum Core)

Parameter	Value	Value
Asset	weETH	WETH
Collateral	Yes	No
Borrowable	No	Yes
Max LTV	93%	-
Liquidation Threshold	95%	-
Liquidation Penalty	1.00%	-

weETH/WETH E-Mode (Arbitrum)

Parameter	Value	Value
Asset	weETH	WETH
Collateral	Yes	No
Borrowable	No	Yes
Max LTV	93%	-
Liquidation Threshold	95%	-
Liquidation Penalty	1.00%	-

weETH LT&LTV

Asset	Chain	Current LTV	Recommended LTV	Current LT	Recommended LT
weETH	Base	72.5%	75%	75%	77%
weETH	Arbitrum	72.5%	75%	75%	77%

wstETH/ WETH E-Mode (Base)

Parameter	Value	Value
Asset	wstETH	WETH
Collateral	Yes	No
Borrowable	No	Yes
Max LTV	93.5%	93.5%
Liquidation Threshold	95.5%	95.5%
Liquidation Penalty	1.00%	1.00%

wstETH/ WETH E-Mode (Arbitrum)

Parameter	Value	Value
Asset	wstETH	WETH
Collateral	Yes	No
Borrowable	No	Yes
Max LTV	93.5%	93.5%
Liquidation Threshold	95.5%	95.5%
Liquidation Penalty	1.00%	1.00%

wstETH/ WETH E-Mode (Ethereum Core)

Parameter	Value	Value
Asset	wstETH	WETH
Collateral	Yes	No
Borrowable	No	Yes
Max LTV	93.5%	93.5%
Liquidation Threshold	95.5%	95.5%
Liquidation Penalty	1.00%	1.00%

Disclaimer

Chaos Labs has not been compensated by any third party for publishing this ARFC.

Copyright

Copyright and related rights waived via CC0

[LlamaRisk]

Summary

LlamaRisk supports these changes. The proposed risk parameter adjustments are rational, given the justifications provided by ChaosLabs. Moreover, we see minimal incremental risks associated with introducing new Liquid E-Modes. This update will allow for a fine-grained risk control for the LST yield leveraging strategy that is dominant on Aave.

weETH

LTV Adjustments

The main use case for weETH collateral on Arbitrum and Base is LST yield leverage, where WETH is mainly borrowed. On Arbitrum, 97.5% of these borrows are made in the ETH Correlated E-Mode, which has an optimized max LTV of 93%. As for Base, 99.7% of WETH borrows using weETH as collateral are made in the ETH Correlated E-Mode with a max LTV of 90%. Most loans in the ETH Correlated E-Mode are at 1-1.1 health ratios.

Arbitrum

Source: ChaosLabs Aave Risk Analytics Dashboard, March 13th, 2025

Base

Source: ChaosLabs Aave Risk Analytics Dashboard, March 13th, 2025

Notably, the proposed LTV adjustments would not impact the utility of the LST yield looping on these markets. However, increasing LTV could improve the capital efficiency of weETH price leveraging on Aave’s Arbitrum and Base markets as this strategy represents a very small portion of weETH-backed loans (~1.5%). On Aave’s Core market, where the max LTV of weETH is 77.5%, this strategy represents a slightly larger portion of the loans (2.5%). If the negative market conditions persist and stablecoin rates on Aave continue to decrease, this strategy could become more apparent.

Source: Aave Dashboard, March 13th, 2025

Considering these observations and the sustained stability of the liquidity on both Arbitrum and Base, the proposed increases in weETH LTV are rational. They would bring the utility of weETH collateral outside E-Modes closer to that of Aave’s Core market.

weETH/WETH E-Mode

The current general ETH Correlated E-Mode allows up to 93% LTV on Core and Arbitrum while offering a 90% max LTV on Base. Creating a new specialized weETH/WETH E-Mode on these 3 markets would unify the maximal LTV while offering a lower Liquidation Bonus on Base.

Source: Aave Dashboard, March 13th, 2025

Creating a weETH/WETH Liquid E-Mode allows the enactment of fine-grained risk parameter adjustments and, therefore, more collateral efficiency for specific strategies. With the introduction of Umbrella, these changes will help to set the liquidity coverage targets more efficiently and thus allow to push the LTV parameters further, as proposed in a subsequent Risk Stewards publication.

wstETH

The amount of wstETH supply on Arbitrum and Base is much smaller than weETH. While most activity is LST yield looping, more diverse use cases of collateral and increased stablecoin borrowing are observed in these markets. In addition, wstETH collateral usage on Aave’s Core market is also more distributed, with the borrowing of WETH in ETH Correlated E-Mode representing 50% of all borrows. Similarly to weETH, most loans backed by wstETH in the ETH Correlated E-Mode are at 1-1.1 health ratios.

Core

Source: ChaosLabs Aave Risk Analytics Dashboard, March 14th, 2025

Arbitrum

Source: ChaosLabs Aave Risk Analytics Dashboard, March 14th, 2025

Base

Source: ChaosLabs Aave Risk Analytics Dashboard, March 14th, 2025

wstETH/WETH E-Mode

For wstETH, the ETH Correlated E-Mode is the sole setup used to borrow more efficiently. Introducing a separate Liquid E-Mode for wstETH/WETH looping would push the maximal LTV to 93.5%, which is 0.5% higher than the current max LTV on Arbitrum/Core and 3.5% higher than the current max LTV on Base. Once again, implementing this separate E-Mode would allow further optimizations, enabling more yield leverage for the users. The fine-grained setup isolates risks and pushes for more aggressive LTV levels but remains to be covered via the Umbrella.

Further Risk Considerations

This proposal sets a precedent for moving LTV for the LST yield looping strategy on Aave to higher levels than currently. This strategy’s correlated nature of collateral and borrowed assets is the underlying reason for high LTVs. This means that the sole reason for liquidations could be the unlikely event of LST slashing. The potential slashing risk is estimated in more detail as part of our Ethereum Validator Slashing Simulator.

While the loan-to-value buffer has been kept higher, introducing Umbrella’s automated risk coverage will allow growth managers to propose more utility and lower this buffer. The upcoming aim of risk providers will be to ensure that optimal insurance levels mitigating increased risks are set.

Disclaimer

This review was independently prepared by LlamaRisk, a community-led decentralized organization funded in part by the Aave DAO. LlamaRisk is not directly affiliated with the protocol(s) reviewed in this assessment and did not receive any compensation from the protocol(s) or their affiliated entities for this work.

The information provided should not be construed as legal, financial, tax, or professional advice.